Status of Hybrid Wing Body Community Noise Assessments · Status of Hybrid Wing Body Community...

26
National Aeronautics and Space Administration Status of Hybrid Wing Body Community Noise Assessments Russell H. Thomas, Casey L. Burley, and Erik D. Olson NASA Langley Research Center, Hampton, VA USA Presented at the AIAA Aerospace Sciences Meeting Special ERA Session Orlando, Florida January 6, 2011 www.nasa.gov 1

Transcript of Status of Hybrid Wing Body Community Noise Assessments · Status of Hybrid Wing Body Community...

National Aeronautics and Space Administration

Status of Hybrid Wing Body

Community Noise Assessments

Russell H. Thomas, Casey L. Burley, and Erik D. Olson

NASA Langley Research Center, Hampton, VA USA

Presented at the AIAA Aerospace Sciences Meeting

Special ERA Session

Orlando, Florida

January 6, 2011

www.nasa.gov

1

Introduction

Acknowledgments:

Mr. Craig Nickol, Dr. Frank Gern, Mr. Jeff Berton, NASA

PAA Experimental Data on separate task: Dr. Michael Czech, Dr.

Leon Brusniak, Mr. Ronen Elkoby and the LSAF Team, The

Boeing Company

Mr. John Rawls Jr., Lockheed Martin Engineering Services

PAA Definition: Aeroacoustic effects associated with the integration of

the propulsion and airframe systems (acoustic and flow interactions)

2

Outline

• NASA N+2 Noise Goal Roadmap

HWB with BPR 7 Turbofan (ref: AIAA 2010-3913 Thomas, Burley, and Olson)•

• Assessment Process Including PAA Experiment

HWB Configurations

System Noise Impacts

Summary & Future Directions – HWB with UHB Turbofan

• Current Study In Progress – HWB with Open Rotor

• LSAF PAA Experiment

Assessment Process•

• Summary

3

NASA N+2 Acoustics Goal Roadmap

Initial System Noise

Assessment, sets

Stage 4 –42dB Goal

Simple Shielding

Experiment

Basic Concept

Selection > Hybrid

Wing Body (HWB)

Broaden Technology Path,

Refined Methods,

Increasing Confidence

Industry Partnerships Key

Pathfinding

Technology Development

Maturation

HWB Aeroacoustic

Experiments:

• Large-scale, high fidelity

integrated HWB aircraft systems

• Low Noise Levels

Multi-Disciplinary HWB

Aircraft Concept Design:

• Absence of prior validation and

acoustic prediction methods

HWB Aircraft System Prediction Methods:

• Interaction effects methods

• Validation Experiments

• Vehicle Definition and Computational Capabilities

2009 Preliminary

with LSAF Data

4

System Noise

Assessments

AIAA 2010-3913

with LSAF Data

2005 Pathfinding with

Limited Data

2012 High Fidelity

with 14 by 22

2011 with Open Rotor

LSAF PAA data

Engine Options for HWB:

• BPR 7 Turbofan

• BPR 10 Turbofan

• Ultra High BPR Turbofan

• Open Rotor Engine Types

Review 2010 Assessment - Aircraft Models and

Framework

Both aircraft use equivalent technology levels and are sized for

same payload, same 7500 NM mission, meet FAA airworthiness

standards, and same GE90-like engine used on both aircraft

SOA HWB

5

Aircraft System Noise Prediction Method

• NASA Aircraft Noise Prediction Program (ANOPP-Lv 27)

• SOA and HWB flight definition from FLOPS

GE90-like relative engine noise sources match data (ref. Gliebe, 2003)

Total SOA prediction calibrated to match EPNL data for this aircraft

ANOPP

FLight OPtimization System (FLOPS)Aircraft Flight Definition

Numerical Propulsion System

Simulation (NPSS)

Source Noise

Jet: ST2JET

Fan: Boeing method

Duct Treatment: GE

Core: SAE

Airframe: Boeing, Fink (t.e.)

Experimental

Data from

LSAF (ref AIAA

2010-3912)

PAA Effects

Propagation & Noise Metrics

EPNL predicted at FAR 36 locations6

Flight Path Profiles

7

Approach Takeoff

Technology and Experimental Data for Key PAA Effects

HWB Experiment Improves Basic Understanding of Aeroacoustic Sources and Parameters:

Effects from Verticals and Elevons

Jet-Airframe shielding including source modification

Broadband point source shielding with flow effect

Ref: Czech, Thomas, and Elkoby, “Propulsion Airframe Aeroacoustics Integration Effects for a Hybrid Wing Body Aircraft Configuration,” AIAA 2010-3912

Effect of Pylon Orientation Relative to Observer

“Crown Pylon” position“Keel Pylon” position

SPL difference between crown pylon and keel position (from ref AIAA 2010-3912)

• Effect of pylon orientation increases with power setting• Azimuthal orientation of pylon has up to 8dB effect in aft arc

Effect of pylon on key jet noise source included through experimental information

Crown rel. to Keel at Cutback Crown rel. to Keel at Sideline

20 25 30 3510*log(f), Hz

20 25 30 3510*log(f), Hz

Emis

sion

Ang

le, d

eg

SPL

9

PAA Technology Effects on System Noise Levels

+2D -1D

Baseline, Engines

1D downstream of

trailing edgeSimple Shielding,

engines move 2D

upstream

10

Jet noise changes with chevronsSpectral changes with Chev1

Mild designPeak source location

changes

• Chevrons relocate peak sources towards th

nozzle exit except at very low frequencies

• Movement of sources with chevrons is

favorable for shielding

e

Spectral changes with Chev2

Aggressive design

High frequency

increase

Low frequency

reduction

SPL

SPL

11

Shielding Effectiveness

Normalized jet noise EPNdB vs X/D

Sideline conditionCutback condition

• Jet noise EPNdB varies significantly as a function of engine location. Isol

nozzle is the reference.

• The baseline nozzle with pylon offers reductions of ~1 to 2dB of EPNdB

• Shielding effectiveness significantly enhanced with the chevron nozzle

• Jet noise EPNdB decreased by up to 5dB at x/D=2 and cutback power

ated

Increasing slope

means more noise

reduction with the

same surface area

12

PAA Technology Effects on System Noise Levels

Baseline, engines

1D downstream of

trailing edge

Simple Shielding,

engines move 2D

upstream

2D, chevrons to

reduce jet source &

increase shielding

effectiveness

Vertical

Pylon at

Keel

Chevrons

13

PAA Technology Effects on System Noise Levels

Favorable

directivity of

jet from

strong effect

crown pylon

Acoustic liner

added to

crown pylon

for aft fan

attenuation

Pylon at

Crown

Keel PylonCrown Pylon

Pylon rotated from keel to

crown position14

PAA Technology Effects on System Noise Levels

Baseline, Engines

1D downstream of

trailing edge

Simple Shielding,

Engines move 2D

upstream

Chevrons reduce

source & increase

shielding

effectiveness (same

area, more noise

reduction)

Active Pylon

adds more jet

shielding

effectiveness,

elevons add

more aft fan

shielding

Favorable

directivity of

jet from

strong effect

crown pylon

Acoustic liner

added to

crown pylon

for aft fan

attenuation

Projection from more

advanced crown

pylon, chevrons, quiet

landing gear

15

EPNL Impacts of HWB Configurations

• HWB configuration results in a new distribution between 3 cert points

Technology reduces all 3 cert points without changing basic distribution•

16

Perspective on 42.4 dB Cumulative Level

From AIAA 2009-3144

Note: Included a BPR 16 engine

SOA HWB

Technology that

reduces noise sources

and relocates sources

to be more effectively

shielded

Simple shielding from

moving engines

upstream on aircraft

Lower noise of baseline

HWB from: simple

shielding of inlet noise,

lower approach speed,

absence of flap noise,

flight path effects

An integrated systems approach key to achieving NASA’s

N+2 goal of 42 dB17

Sound Exposure Level (SEL) Contour

SOA

HWB

Impact is a 66% reduction in ground area

For a simulated

certification procedure

(FAR 36) takeoff and

landing

18

Summary of 2010 Assessment of HWB with BPR 7 (from ref AIAA 2010-3913 Thomas, Burley, and Olson)

• A rigorous HWB system noise assessment with key elements:

– NASA ANOPP system noise method

NASA updated HWB aircraft model and flight path

Boeing/NASA PAA LSAF (2009) experimental results

• 42.4 dB cumulative assessed on the HWB with relatively near term

technologies:

– Existing GE90-like engine (BPR 7)

PAA chevron nozzle and crown pylon technology configurations

Acoustic liner applied on the crown pylon

Quiet landing gear technology

Reduced approach flight speed

Results in higher confidence assessment

compared to earlier pathfinding assessments

19

Future Directions – HWB with UHB Turbofan

Better suppression map with more realistic fan noise simulation

Flight path and aircraft model

Maturation of specific PAA and aircraft system technology targeted for noise reduction

Assess with new

technology higher

BPR engines

Keel pylon with weak

acoustic effect

Crown pylon with

strong acoustic effect

Acoustic liner on the

crown pylon inside duct

Winglets

instead of

inboard

verticals

PAA chevrons on

fan nozzle and

core nozzle

Quiet landing gear

(fairings, spoilers,

etc) (not shown)

Include elevon

shielding

Critical step toward higher fidelity HWB aeroacoustic capabilities

20

Assessment of HWB with Open Rotor

Boeing R&T Image

Open Rotor Isolated Engine Fuel Burn Reduction

Promise….with a Known Noise Challenge

Is an integrated HWB/Open Rotor aircraft system another

solution to meet the ERA goals simultaneously?

ERA Goal

N+1 = 2015**

Technology Benefits Relative

to a Single Aisle Reference

Configuration

N+2 = 2020**

Technology Benefits Relative

to a Large Twin Aisle

Reference Configuration

N+3 = 2025**

Technology Benefits

Noise

(cum below Stage 4) -32 dB -42 dB -71 dB

LTO NO Emissionsx

(below CAEP 6) -60% -75% better than -75%

Performance:

Aircraft Fuel Burn-33% -50% better than -70%

exploit metro-plex* conceptsPerformance:

Field Length -33% -50%

**Technology Readiness Level for key technologies = 4-6. ERA will undertake a time phased approach, TRL 6 by 2015 for “long-pole” technologies

* Concepts that enable optimal use of runways at multiple airports within the metropolitan area

21

NASA Open Rotor System Noise Assessment

Process Elements

From AIAA 2009-3144

Note: Included a BPR 16 engine

SOA HWB BPR 7

Turbofan

HWB

Open Rotor

TBD

1. Modernized Open Rotor

Engine Model from GRC

2. Aircraft Model from Boeing

R&T task (Frank Gern TM)

3. Results from NASA/Boeing Open

Rotor PAA Experiment

4. System Noise Assessment (Thomas, Burley,

Olson, Gern, et al):

certification points

include technology options from experiment

flight path variables

Rigorous systems noise assessments a key to

achieving NASA’s N+2 goal of 42 dB

simultaneously with other goals

22

NASA/Boeing Open Rotor PAA Experiment

Experiment of open rotor PAA effects for both HWB and Tube-and-

Wing aircraft types in Boeing’s LSAF completed November 15, 2010Dr. Michael Czech, Boeing PI & Dr. Russ Thomas, NASA TM

Objectives:• Measure PAA effects• Study options for increasing shielding effectiveness• Acquire data for system noise assessment and prediction methods

Instrumentation:• Far Field Microphones

Near Field Mic TraverseFlow Field SurveyPhased Array TraverseSurface Unsteady Pressure

Funded by the NASA Environmentally Responsible Aviation

Project, Dr. Fay Collier, Project Manager

23

NASA/Boeing Open Rotor PAA Experiment

Experimental Parameter Summary:

rotor speed variation

wind tunnel Mach variation

rotor to airframe relative position, axial and vertical

off-center and centerline positions

inboard verticals, size and cant angle

elevon deflection

24

Summary

• ANOPP System Noise Assessment Process for HWB Aircraft Assembled

• Engine System Model

Aircraft System Model

PAA Experimental Data for Key Aircraft Integration Effects

• Rigorous System Noise Assessment of HWB with BPR 7 Turbofan Completed in 2010

• Technology Path Developed

42 dB Assessed Level with High Confidence on Critical Noise Sources•

• Leads to High Fidelity 14 X 22 N2A HWB Experiment and Validation in 2012

Key Elements in Progress Toward Assessment of HWB with Open Rotor•

• Engine System Model

Aircraft System Model

PAA Open Rotor Experiment

PAA Open Rotor Data Analysis

ANOPP Based System Noise Process

– In Progress by GRC Systems Team

In Progress on Boeing R&T task

Completed on Boeing Task

Initiated

In Progress

• –

• –

• −

• –

For HWB aircraft concept, there has been rapid progress in

technology and assessments to meet the noise goal of the N+2 goals

25

26Your Title Here

26